From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2)
Description of problem:
Mailman is already translated to Japanese.But, the encoding is not
UTF8.It encoded by EUC-JP.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.Run a command;/var/mailman/bin/mmsitepass
Actual Results: The command put EUC-JP encoding character.
Expected Results: The command put UTF8 encoding character.
With nkf ,it is easy to correct this problem.
Since I'm not familar with Japanese translations it would help to
explain what nkf is and how it's easy to fix.
This feels like an upstream issues that needs addressing by the
mailman team rather than Red Hat's packaging. If you can help by
providing an exact fix or patch so I'm not doing this blindly I'll fix
it in our package and shepard it upstream.
Nkf is a character encoding conversion filter.
So, it can convert EUC-JP to UTF8.
Created attachment 104450 [details]
UTF-8 Encording ja .po file
This file made by follow command;
$nkf -w mailman.po >mailman.po.new
The file header still says EUC-JP, which is easily corrected.
(btw, the msgfmt tool converts .po files and automatically corrects
To be completely correct, the output should be always converted
to the LC_CTYPE character set. The python gettext interface
doesn't do this, unlike the C one; Mailman/i18n.py contains
an attempt to rectify this, but it doesn't work.
As a first guess, changing _translation.gettext() to
_translation.ugettext() in Mailman/i18n.py:_() should do what
we want, but I haven't tested it at all.
Patch against mailman-2.1.5-28 submitted upstream at
The charset for annals by Mailman (iso-8859-1) is inconsistent with the one by
Fedora Core 3 is now maintained by the Fedora Legacy project for security
updates only. If this problem is a security issue, please reopen and
reassign to the Fedora Legacy product. If it is not a security issue and
hasn't been resolved in the current FC5 updates or in the FC6 test
release, reopen and change the version to match.
Created attachment 159222 [details]
updated patch from cvs that adds missing replacements
I ran into another place where _( needs to be changed to C_(. Attached is an
update to the mailman-2.1.9-ctypo.patch.
Created attachment 172374 [details]
updated patch from cvs that adds more missing replacements
Tomas, you'll probably hate me for reopening this a day after you closed it...
but I just ran into another place where _( needs changed. The attached patch
updates the ctypo patch from the deve branch to add some fixes for the
bin/withlist script. If I had more time to play with mailman, I'd do a more
thorough search for any remaining missing replacements. As it is, I just fix
them as I find them. :(
If you'd prefer me to open new bugs for any other missing replacements instead
of adding to this one, just let me know. Thanks!
Thanks Todd. I'm glad you found those errors because it made me run "grep -r
'[^C|_]_(' * | less" in Mailman sources and must say that it was an upleasant
surprise... Got to find out which of those occurrences have to be fixed. It
looks that bin/newlist and bin/update have some forgotten _( but I was not brave
enough to examine the rest of the list yet. But I'll have to do that...
I've fixed (hopefully) all the places where I thought it had sense to do that.
I'm afraid to fix something that is not reported to be broken... Feel free to
reopen if you find any problems.